|Publication number||US7313551 B2|
|Application number||US 10/486,768|
|Publication date||Dec 25, 2007|
|Filing date||Aug 9, 2002|
|Priority date||Aug 10, 2001|
|Also published as||EP1419482A2, US20050049828, WO2003015029A2, WO2003015029A3|
|Publication number||10486768, 486768, PCT/2002/2849, PCT/FR/2/002849, PCT/FR/2/02849, PCT/FR/2002/002849, PCT/FR/2002/02849, PCT/FR2/002849, PCT/FR2/02849, PCT/FR2002/002849, PCT/FR2002/02849, PCT/FR2002002849, PCT/FR200202849, PCT/FR2002849, PCT/FR202849, US 7313551 B2, US 7313551B2, US-B2-7313551, US7313551 B2, US7313551B2|
|Original Assignee||William J. Brown|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (47), Non-Patent Citations (24), Classifications (18), Legal Events (7)|
|External Links: USPTO, USPTO Assignment, Espacenet|
The invention relates to a processing device for an automatic perception system. The device comprises a set of histogram calculation units called spatio-temporal neurons (STN), preferably auto-adaptive and possibly having anticipation and learning capacity. They are more particularly intended for image perception and processing. A further object of the invention is a method therefor.
Devices for processing images in real time, for recognizing, localizing and/or extracting their contexts, of objects corresponding to certain criteria are known. The selection criteria can be extremely varied; it can be a question of speed, of a shape, a color or a combination of these criteria.
These devices can be utilized for facilitating the understanding of a scene or of a phenomenon by an observer or for controlling an automatism or automatic operation suing the information extracted in this fashion.
This type of device is described, for example, in the following publications FR 2.611.063 and WO 98/05002.
These devices implement a spatial and temporal processing unit that, receiving a S(PI) video signal, produces a certain number of parameters for each pixel. For example, this is the speed V, the direction DL, a certain time constant CO, and a binary VL validation parameter in addition to a delayed video signal VR and different frame, line and pixel synchronization signals grouped under the designation F.
In such devices, the interest in creating histograms of these parameters make it possible to structure, manipulate and exploit static information.
The purpose of these image processing devices is to provide at output a signal S′(t) that carries for each pixel meaningful information on the results of the application of the recognition or selection criteria. These criteria are predefined or elaborated by the imaging processing methods and devices themselves.
In particular, one such devices is known from the description in the WO 98/05002 patent application cited hereinbefore. In one such device, a set of STN calculation modules as represented in
The object of the invention is to provide an automatic perception device that is improved relative to the prior art and which makes possible more flexible processing of the responses of the calculation units and their functional and hierarchical grouping, while reducing the number of electronic elements that must be used with respect to the preceding solution. The invention thus relates to a processing device for an automatic perception system utilizing STN calculation modules that receive data from a data bus and which are interconnected via a backannotation bus.
According to the invention, the modules are grouped into hierarchized sets comprising the set of order 0 formed by a single module, the set of order 1 formed by the combining of a plurality of order 0 sets, the sets of order P greater than 1 formed by the combination of lower order P−1 sets, the hierarchized sets of a given order P sharing a backannotation bus and, the interconnection of the backannotation bus between a lower order P and an order greater P+1 is done by means of an interconnection module according to an interconnection configuration as a function of selection parameters.
The present invention also relates to the characteristics that will become apparent in the course of the following description and which must be considered individually or in terms of their possible technical combinations:
By now considering the device invention starting with its lowest order, order 0; that is, the STN calculation modules and moving up towards the greater orders, the present invention for order 0 is a processing device for an automatic perception system comprising at least one set of N STN calculation modules, a data bus, a backannotation bus with N backannotation lines, the modules being interconnected and each of the modules comprising one histogram calculation unit performing a calculation as a function of a parameter selected in a data selection unit arranged between the data bus and the calculation unit, said parameter being in addition taken into account by a classification unit intended to produce a decision-making signal according to the result of the calculation unit, the decision-making signal being sent over the backannotation bus, the calculation of the calculation unit being validated by a signal originating from a backannotation selection unit arranged between the backannotation bus and the calculation unit, wherein each of the modules comprises in addition a compound term generator receiving on input, on the one hand, a decision-making signal coming from the classification unit and, on the other hand, N−1 lines of backannotation signals, the backannotation signal from the module under consideration being excluded, the generator producing on output a backannotation signal sent on the backannotation bus as a function of the selection parameters. In one preferred embodiment, the backannotation selection unit is a compound term generator.
By now considering the greater order; that is, order 1, corresponding to a group formed by a set of STN calculation modules on the same backannotation bus and the orders beyond 1; that is, sets of groups, the set of N modules is divided up into a plurality of hierarchized sub-sets, each of the sub-sets forming a hierarchical functional group of order P>=1 on a P group backannotation bus, the interconnection of the backannotation bus between a first order group P with a second greater order group P+1 is done by means of an interconnection module.
The invention also relates to a method of processing for an automatic perception system utilizing STN calculation modules that receive data from a data bus and which are interconnected via a backannotation bus.
According the method invention, using a device according to any one of the aforementioned characteristics, possibly combined, the modules are grouped into hierarchical sets comprising the set of order 0 formed by a single module, the order 1 set formed by the combination of a plurality of lower order 0 sets, the order P sets greater than 1 formed by the combination of sets of the lower order P−1, the hierarchized sets of a given order P sharing a backannotation bus and one interconnects the backannotation buses between a lower order P and a greater order P+1 by means of an interconnection module according to the interconnection configuration as a function of the selection parameters.
In one embodiment of the method, one realizes parametrizable inversions of the input signals E0, E1, . . . En for producing signals E′0, E′1, . . . E′n and as a function of the parameters, an output signal out is produced resulting from either the AND combination with:
In virtue of the invention, a functional hierarchized structure, performing calculations, producing results in the form of decisions sent over a backannotation bus can easily realized. This structure is arborescent between/among the different group levels.
Each of the groups appears for the higher group as a (super) module, this (super) module being consistently complex in the case of P>1 in that it groups one, and preferably, a plurality of modules. The sub-sets are not disjoint, since it is possible that one greater order group may be formed by the combination of one, and preferably, a plurality of lower order groups. It is thus possible to functionally group according to parameters, sub-sets of modules in order to form a functional (super) module and it is likewise possible to functionally group according to parameters a plurality of these (super) modules. The calculations within the system can thus be hierarchized and within each hierarchical level, the number and the calculations of the modules, (super) modules, etc., in terms of the hierarchical level, can be parametrizable. The invention also makes it possible to optimize the size of the bus with respect to a prior art device, because using the invention the buses between/among the different improved modules and/or the super modules of a plurality of hierarchized groups convey the data resulting from parametrable combinations contrary to the prior art.
The invention will now be described in more detail with reference to the annexed drawings, wherein:
It is understood that it is then possible to construct a structure comprising even higher orders by interconnection of the backannotation bus by means of interconnection modules. In these structures, the groups of different order or of the same order can be of the same or different size. It is also possible to more easily hierarchically structure a perception device and method, for example a first group being capable of specializing in a right eye, another for a left eye, yet another for the nose, etc. for the different elements of a face and said groups being interconnected for a greater order corresponding to recognition of a face and in a general fashion, any memorized formal shape generated by previous learning or a pre-defined conceptualization of the shape.
In a variant embodiment (not shown), more than one second compound term generator 100 c is utilized for sending via the lower order backannotation bus more than one backannotation signal.
Preferably, the signal produced by the interconnection module to the greater order Bs bus is not sent on input of the third compound term generator 100 b.
Finally, in virtue of the utilization of the parameter storage registers for the interconnection modules, an hierarchized module set is a memory that can correspond to a particular element that is analyzed in an environment by the automatic perception system. The memory can be defined by the configuration of the selection parameters of the interconnection modules within the set. In an enriched environment; that is, comprising numerous elements, the device can make it possible to memorize in terms of the configuration of the selection parameters by performing more functional combinations between the bus signals of different orders and thus between the sub-sets of modules.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US4783828||Jun 2, 1986||Nov 8, 1988||Honeywell Inc.||Two-dimensional object recognition using chain codes, histogram normalization and trellis algorithm|
|US5008946||Sep 9, 1988||Apr 16, 1991||Aisin Seiki K.K.||System for recognizing image|
|US5088488||Dec 22, 1989||Feb 18, 1992||Medtronic, Inc.||Method and apparatus for implementing histogram storage and trend analysis in a medical stimulator|
|US5109425||Sep 30, 1988||Apr 28, 1992||The United States Of America As Represented By The United States National Aeronautics And Space Administration||Method and apparatus for predicting the direction of movement in machine vision|
|US5163095||Feb 26, 1991||Nov 10, 1992||Toa Medical Electronics Co., Ltd.||Processor for extracting and memorizing cell images, and method of practicing same|
|US5278921||May 21, 1992||Jan 11, 1994||Fuji Photo Film Co., Ltd.||Method of determining exposure|
|US5359533||Feb 11, 1993||Oct 25, 1994||Asea Brown Boveri Ltd.||Method for determining a frequency/time profile of hits, and device for carrying out the method|
|US5384865||Jun 1, 1992||Jan 24, 1995||Eastman Kodak Company||Adaptive, hybrid median filter for temporal noise suppression|
|US5488430||Dec 22, 1992||Jan 30, 1996||Goldstar Co., Ltd.||Motion detection and estimation apparatus and method thereof|
|US5565920||May 26, 1995||Oct 15, 1996||The Trustees Of Princeton University||Method and apparatus for video data compression using temporally adaptive motion interpolation|
|US5592226||Jan 26, 1994||Jan 7, 1997||Btg Usa Inc.||Method and apparatus for video data compression using temporally adaptive motion interpolation|
|US5592237||Nov 4, 1994||Jan 7, 1997||Infimed, Inc.||High resolution image processor with multiple bus architecture|
|US5625717||Jun 22, 1993||Apr 29, 1997||Mitsubishi Denki Kabushiki Kaisha||Image processing device for processing grey level images|
|US5694495||Nov 13, 1995||Dec 2, 1997||Mitsubishi Denki Kabushiki Kaisha||Image sensing device and image sensing method|
|US5712729||May 15, 1995||Jan 27, 1998||Olympus Optical Co., Ltd.||Artificial retina cell, artificial retina and artificial visual apparatus|
|US5774581||Feb 3, 1995||Jun 30, 1998||U.S. Philips Corporation||Device for segmenting a discrete assembly of data|
|US5793888||Jan 29, 1997||Aug 11, 1998||Massachusetts Institute Of Technology||Machine learning apparatus and method for image searching|
|US6256608||Jun 30, 1998||Jul 3, 2001||Microsoa Corporation||System and method for entropy encoding quantized transform coefficients of a signal|
|US6304187 *||Jul 22, 1999||Oct 16, 2001||Holding B.E.V. S.A.||Method and device for detecting drowsiness and preventing a driver of a motor vehicle from falling asleep|
|US6486909 *||Jul 22, 1997||Nov 26, 2002||Holding B.E.V.||Image processing apparatus and method|
|US6597738||Aug 27, 1999||Jul 22, 2003||Hyundai Curitel, Inc.||Motion descriptor generating apparatus by using accumulated motion histogram and a method therefor|
|US6717518 *||Jan 15, 1999||Apr 6, 2004||Holding B.E.V.S.A.||Method and apparatus for detection of drowsiness|
|US6959293 *||Feb 23, 2001||Oct 25, 2005||Holding B.E.V.S.A.||Method and device for automatic visual perception|
|US7043465 *||Jun 8, 2001||May 9, 2006||Holding B.E.V.S.A.||Method and device for perception of an object by its shape, its size and/or its orientation|
|US7080054 *||Jul 16, 2004||Jul 18, 2006||Idaho Research Foundation, Inc.||Biomimic artificial neuron|
|US7136842 *||Jun 8, 2001||Nov 14, 2006||Holding B.E.V. S.A.||Method and device for adapting a signal for the detection of the movement of an object|
|US20020071595 *||Feb 23, 2001||Jun 13, 2002||Patrick Pirim||Image processing apparatus and method|
|US20020101432||Jun 22, 1999||Aug 1, 2002||Kazuhiro Ohara||Histogram-based intensity expansion|
|US20020120594 *||Jun 8, 2001||Aug 29, 2002||Patrick Pirim||Method and device for perception of an object by its shape, its size and/or its orientation|
|US20020156753 *||Jun 8, 2001||Oct 24, 2002||Holding B.E.V.S.A.||Method and device for adapting a signal for the detection of the movement of an object|
|US20020169732||Feb 23, 2001||Nov 14, 2002||Patrick Pirim||Method and device for automatic visual perception|
|US20030067978 *||Jul 3, 2002||Apr 10, 2003||Holding B.E.V.||Image processing apparatus and method|
|US20030152267 *||Feb 23, 2001||Aug 14, 2003||Patrick Pirim||Automatic perception method and device|
|US20050049828 *||Aug 9, 2002||Mar 3, 2005||B.E.V.S.A.||Processing device and method for an automatic perception system|
|US20050102247 *||Jul 16, 2004||May 12, 2005||Wells Richard B.||Biomimic artificial neuron|
|EP0046110A1||Aug 3, 1981||Feb 17, 1982||Michel Urien||Device for obtaining a time-distance histogram of successive events|
|EP0380659A1||Oct 2, 1987||Aug 8, 1990||Kabushiki Kaisha Komatsu Seisakusho||Method of recognizing linear pattern|
|EP0394959A2||Apr 24, 1990||Oct 31, 1990||Fraunhofer-Gesellschaft Zur Förderung Der Angewandten Forschung E.V.||Image analysis method|
|FR2611063A1||Title not available|
|FR2751772A1||Title not available|
|JPH06205780A||Title not available|
|WO1998005002A1||Jul 22, 1997||Feb 5, 1998||Carlus Magnus Limited||Method and device for real-time detection, location and determination of the speed and direction of movement of an area of relative movement in a scene|
|WO1999036893A1||Jan 15, 1999||Jul 22, 1999||Holding B.E.V. S.A.||Method and apparatus for detection of drowsiness|
|WO1999036894A1||Jan 14, 1999||Jul 22, 1999||Carlus Magnus Limited||Method and device for detecting drowsiness and preventing a driver of a motor vehicle from falling asleep|
|WO2000011609A1||Jan 25, 1999||Mar 2, 2000||Holding B.E.V. S.A.||Image processing apparatus and method|
|WO2000011610A1||Aug 25, 1998||Mar 2, 2000||Holding Bev S.A.||Image processing apparatus and method|
|WO2001063557A2||Feb 23, 2001||Aug 30, 2001||Holding B.E.V. S.A.||Automatic perception method and device|
|1||"Elecktronik Revue" ER, Eine Elsevier-Thomas-Publikation, Jahrgang 8, Marz 1997, NR.3, ISSN0939-1134.|
|2||"Realiser Un Decodeur Pour TV Numberique", Electronique, Le Mensuel Des Ingenieurs De Conception, No. 66, Janvier 1997 (with translation).|
|3||"Un Processor de Perception Visuelle", LehAUG pARLEUR, 25F Des solutions électroniques pour tous, N° 1856, 15 janvier 1997 (with translation).|
|4||*||A recurrent spatio-temporal neuron: advantage of potential build-up Clouse, R.L.; Jun Liu; Waldron, M.B.; [Engineering in Medicine and Biology, 1999. 21st Annual Conf. and the 1999 Annual Fall Meeting of the Biomedical Engineering Soc.] BMES/EMBS Conference, 1999. Proceedings of the First Joint vol. 1, Oct. 13-16, 1999 pp. 395 vol. 1.|
|5||*||A recurrent spatio-temporal neuron: advantage of potential build-up Clouse, R.L.; Jun Liu; Waldron, M.B.;[Engineering in Medicine and Biology, 1991. 21st Annual Conf. and the 1999 Annual Fall Meeting of the Biomedical Engineering Soc.] BMES/EMBS Conference, 1999. Proceedings of the First Joint vol. 1, Oct. 13-16, 1999 p. 395 vol. 1.|
|6||Adsani, Nabeel Al, "For Immediate Release The Generic Visual Perception Processor", Oct. 10, 1997, p. 1.|
|7||*||Computer lipreading for improved accuracy in automatic speech recognition Silsbee, P.L.; Bovik, A.C.; Speech and Audio Processing, IEEE Transactions on vol. 4, Issue 5, Sep. 1996 pp. 337-351 Digital Object Identifier 10.1109/89.536928.|
|8||Daugman, John G., "Complete Discrete 2-D Gabor Transforms by Neural Networks for Image Analysis and Compression", IEEE Transaction on Acoustics, Speech and Signal Processing, vol. 36, No. 7, Jul. 1988, pp. 1169-1179.|
|9||Electroncique International Hebdo, Dec. 5, 1996-N° 245, "Premier . . . L'oeil". Francoise Gru svelet (with translation).|
|10||Electronique, "Le Mechanisme de la Vision Humaine Dans Le Silicium", Electronique Le Mensuel Des Ingenieurs De Conception, No. 68, Mars 1997, ISSN 1157-1151 (with translation).|
|11||Es Professionnels de Linformatique En Entreprise Magazine, "Objectif Securite Des Reseaux", No. 24, Janvier, 1997.|
|12||Groupe Revenu Français, Air & Cosmos Aviation International, "Un Calculateur De Perceoption Visuelle", Hebdomadaire, vendredi 6 décembre 1996, 34 Année, No. 1590, 22F.|
|13||*||Humanoid Head Prototype with Uncoupled Eyes and Vestibular Sensors Ouezdou, F.B.; Alfayad, S.; Pirim, P.; Barthelemy, S.; Intelligent Robots and Systems, 2006 IEEE/RSJ International Conference on Oct. 2006 pp. 2980-2985 Digital Object Identifier 10.1109/IROS.2006.282231.|
|14||Indiveri, Giacomo, et al., "System Implementations of Analog VLSI Velocity Sensors", 1996 IEEE Proceedings of MicroNeuro '96, pp. 15-22.|
|15||Inside the Pentagon's Inside Missile Defense, an exclusive biweekly report on U.S. missile defense programs, procurement and policymaking, "Missile Technology" vol. 3, No. 16-Aug. 13, 1997, p. 5.|
|16||Johnson, Colin, "Vision Chip's Circuitry Has Its Eye Out For You", http://184.108.40.206/wire/news/1997/09/0913vision.html, pp. 1-3.|
|17||*||Learning pulse coded spatio-temporal neurons with a local learning rule Klaassen, A.J.; Dev, A.; Neural Networks, 1991., IJCNN-91-Seattle International Joint Conference on vol. i, Jul. 8-14, 1991 pp. 829-836 vol. 1 Digital Object Identifier 10.1109/IJCNN.1991.155286.|
|18||*||Learning pulse coded spatio-temporal neurons with a local learning rule Klaassen, A.J.; Dev, A.; Neural Networks, 1991., IJCNN-91-Seattle International Joint Conference on vol. i, Jul. 8-14, 1991, pp. 829-836 vol. 1.|
|19||Mallat, Stephane G., "A Theory for Multiresolution Signal Decomposition: The Wavelet Representation", IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 11, No. 7, Jul. 1989, pp. 674-693.|
|20||Revue Trimestrielle Des <<Techniques de Lingenieur>>, "Instantanés Technique" Techniques De L'ingénieur, Mars 1997-N°5 (40F), ISSn 0994-0758.|
|21||Rüedi, Pierre-Francois, "Motion Detection Silicon Retina Based on Event Correlations", 1996 IEEE Proceedings of MicroNeuro '96, pp. 23-29.|
|22||The Japan Times, :"British firm has eye on the future", Business & Technology, Tuesday, Nov. 18, 1997, 4th Edition.|
|23||Tomita, Jr., Alberto, et al., "Hand Shape Extraction from a Sequence of Digitized Gray-Scale Images", IECON '94, 20th International Conference on Industrial Electronics, Control and Instrumentation, vol. 3 of 3, Special Sessions, Signal Processing and Control, pp. 1925-1930.|
|24||Yamada, K., et al; "Image Understanding Based on Edge Histogram Method for Rear-End Collision Avoidance System", Vehicle Navigation & Information Systems Conference Proceedings; (1994), pp. 445 450 published Aug. 31, 1994; XP 000641348.|
|U.S. Classification||706/43, 382/168, 702/180, 706/15|
|International Classification||G06T5/00, G06F17/18, G06G7/00, G06T7/00, G06F15/18, G06E1/00, G06F19/00, G06K9/00, G06E3/00, G06T1/20, G06T1/40|
|Cooperative Classification||G06T2200/28, G06T5/40|
|Aug 26, 2004||AS||Assignment|
Owner name: B.E.V. S.A., LUXEMBOURG
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:PIRIM, PATRICK;REEL/FRAME:015088/0298
Effective date: 20040728
|Feb 1, 2006||AS||Assignment|
Owner name: U.S DISTRICT COURT, SO. DIST. FL, FLORIDA
Free format text: PRELIMINARY INJUNCTION;ASSIGNOR:B.E.V. S.A.;REEL/FRAME:017103/0629
Effective date: 20041213
|Feb 2, 2006||AS||Assignment|
Owner name: WILLIAM J. BROWN, ESQ., TRUSTEE FOR JOSEPH D. HARB
Free format text: ASSIGNMENT BY CLERK U.S. DISTRICT COURT;ASSIGNOR:U.S. DISTRICT COURT, SO. DIST. FLA. (BY U.S. COURTORDER);REEL/FRAME:017109/0726
Effective date: 20050728
Owner name: U.S. DISTRICT COURT, SO. DIST, FL, FLORIDA
Free format text: PERMANENT INJUNCTION FINAL JUDGMENT;ASSIGNOR:B.E.V. S.A.;REEL/FRAME:017109/0185
Effective date: 20050111
Owner name: WILLIAM J. BROWN, ESQ., TRUSTEE FOR JOSEPH D. HARB
Free format text: ASSIGNMENT BY U.S. MARSHALL;ASSIGNOR:U.S. DISTRICT COURT, SO. DIST. FLA. (BY U.S. COURT ORDER);REEL/FRAME:017109/0733
Effective date: 20060111
|Oct 23, 2009||AS||Assignment|
Owner name: IMAGE PROCESSING TECHNOLOGIES LLC, NEW YORK
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:BROWN, WILLIAM J., AS TRUSTEE FOR JOSEPH D. HARBAUGH;REEL/FRAME:023409/0987
Effective date: 20091022
|May 16, 2011||FPAY||Fee payment|
Year of fee payment: 4
|May 5, 2015||AS||Assignment|
Owner name: IMAGE PROCESSING TECHNOLOGIES LLC, FLORIDA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:IMAGE PROCESSING TECHNOLOGIES LLC;REEL/FRAME:035567/0221
Effective date: 20150428
|Jun 10, 2015||FPAY||Fee payment|
Year of fee payment: 8